In a data center, constraints (for example, an upper limit value of power consumption) may be set for a power consumed by an information processing device such as a server in order to suppress power consumption.
The power output by a power supply unit (PSU) of the information processing device fluctuates depending on an operation of an application program executed by a processor such as a central processing unit (CPU). Therefore, depending on the operation condition of the application program, the power consumption of the information processing device exceeds the upper limit value. On the other hand, if the power to be supplied to a hardware device such as the CPU is limited too much, the performance of the information processing device deteriorates.
There has been technology regarding suppression of power consumption of the information processing device in which the maximum power consumption of the CPU is controlled by changing the operation mode of the CPU to suppress power consumption of a personal computer (PC). However, since this technology uses the predicted value as the maximum power consumption, whether the power consumption is properly suppressed depends on the accuracy of the prediction.
There has been technology of controlling the ratio of the power distributed to a plurality of components based on at least one operation condition of the plurality of components to which the power is supplied. However, in this technology, the power consumption of the information processing device may not be appropriately controlled, because it depends on the control of the ratio of the power allocated to the plurality of components.
The followings are reference documents.
[Document 1] Japanese Laid-open Patent Publication No. 2000-172387 and
[Document 2] Japanese Laid-open Patent Publication No. 2016-189109.